Stream: Orders and Observation WG
Topic: Link DiagnosticReports hierarchically
Patrick Werner (Feb 04 2019 at 17:01):
This is a follow up on: GF#17205
We still have the need to have DiagnosticReports which are summarizing several other DiagnosticReports. In the GF you pointed to the existing Extensions.
We are currently considering the usage of: http://build.fhir.org/extension-diagnosticreport-extends.html
Patrick Werner (Feb 04 2019 at 17:02):
Would you agree on extends, or do you think another of the several extensions would be a better fit?
Eric Haas (Feb 04 2019 at 21:47):
have you considered http://build.fhir.org/extension-diagnosticreport-summaryof.html
Patrick Werner (Feb 05 2019 at 15:11):
yes we also considered it. What is the semantical difference between these two options. They seem to overlap a lot.
Andrea Pitkus, PhD, MLS(ASCP)CM, CSM (Feb 05 2019 at 16:37):
Realize the topic has been circling between different WGs. I'd shared with Eric last week, an example of a Diagnostic Report, which integrates results from other lab reports (CBC) that would be reported in DR resource if using FHIR: https://www.mayocliniclabs.com/test-updates/attachment.php?id=52148 Also provided a cytogenetics report as example of genomics testing.
The major issue mentioned on CG calls is around the requirements that each report individually would require the name of the performing lab, patient demographics, etc. However, in an integrated report, each individual report information is often pulled in to the integrated report, but it excludes the patient information from the individual reports in favor of having a header with the patient demographics, performing lab, etc. to avoid unnecessary duplication of fields.
Both types of formats/structure/modeling need to be supported in FHIR DR as they are comprise the major ways narrative reports are communicated worldwide today. It's a broader O&O issue and why Patrick is bringing it forward for O&O to consider. Would there be an upcoming O&O call where it might be discussed Eric or would you have it on the O&O on FHIR call? Would prefer to get Kathy and others to weigh in on how they might support given they report hundreds if not thousands daily in their labs.
Patrick Werner (Feb 06 2019 at 09:36):
The major issue mentioned on CG calls is around the requirements that each report individually would require the name of the performing lab, patient demographics, etc. However, in an integrated report, each individual report information is often pulled in to the integrated report, but it excludes the patient information from the individual reports in favor of having a header with the patient demographics, performing lab, etc. to avoid unnecessary duplication of fields.
i don't see that as an issue in FHIR as these are only references.
Andrea Pitkus, PhD, MLS(ASCP)CM, CSM (Feb 11 2019 at 16:11):
However for US and multiple country regulatory requirements this is an issue as they need to be on each report or represented once on a single integrated report. From what Kevin discussed it didn't seem like this had been resolved. If there are examples it may help clarify.
Eric Haas (Feb 11 2019 at 19:01):
In FHIR context conduction is not assumed and resources must for a complet representation by themselves so "to avoid unnecessary duplication of fields." is not a goal, the resources will by design repeat a lot of information.
Andrea Pitkus, PhD, MLS(ASCP)CM, CSM (Mar 04 2019 at 17:45):
Context is vital to have clinically accurate meaning. I wasn't following your message completely. Are you saying resources must completely represent concepts themselves? If so, then service request and observation and diagnostic report must all be combined? There is important context in lab orders that apply to the results in a number of cases (observations, and thus diagnostic report). It's also vital to know if Ear is a specimen type or specimen source as they are coded differently in SCT, and handled differently in different lab areas (i.e. source is likely a microbiology culture, but as a specimen, it's like anatomic pathology)
If that isn't what you are saying, please clarify.
Last updated: Apr 12 2022 at 19:14 UTC
